A Comparative Study on Automated Detection of Malaria by Using Blood Smear Images

Abstract

Malaria is a parasitic disease or mosquito-borne blood disease. When the mosquito bites a human being, that particular parasite is freed into the human being bloodstream and infects the red blood cells which cause the malaria. We need to understand if the blood-related illness is malaria or not before we provide the right therapy. For this purpose, we must diagnose red blood cells by recognizing or counting red blood cells (erythrocytes). It is very difficult to manually count and recognize infected red blood cells while testing under a microscope by pathologists because maybe it leads to different variations. The current paper gives an overview of the comparison of three different papers with three different techniques used to identify that the red blood cells are infected or not with great accuracy and also to identify which methods are giving best result while performing the diagnosis automatically. With different techniques and methods like Otsu threshold method, global threshold method and classifiers like artificial neural network and support vector machines. All these techniques and methods are related to the diagnosis of the malaria automatically which will reduce the time taken for performing the diagnosis and also it improves the consistency and gives the accurate, rapid result in diagnosis. From the above three methods used, an attempt has been made to finalize the best method from the above three methods.